doctor
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to perform environment checks, including binary location via 'command -v' and executing a local diagnostic script located at '${CLAUDE_PLUGIN_ROOT}/scripts/qmd-list-collections.sh'.
- [DATA_EXFILTRATION]: The skill accesses potentially sensitive information by reading configuration and cache files stored in the user's home directory (~/.cache/qmd/index.sqlite and ~/.config/qmd/index.yml) and project-level Git hooks (.git/hooks/post-commit).
- [EXTERNAL_DOWNLOADS]: The documentation references an external Node.js package '@tobilu/qmd' as a remediation step for missing requirements.
- [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface where values from a project-controlled configuration file are used in shell commands. (1) Ingestion points: The .claude/qmd.json file is read and parsed. (2) Boundary markers: No boundary markers or instructions to ignore embedded commands are present. (3) Capability inventory: The Bash tool is authorized and used to execute grep commands using variables from the configuration. (4) Sanitization: No sanitization or escaping is performed on variables before they are passed to the shell.
Audit Metadata